8 papers · 1 filter
Safer by Diffusion, Broken by Context: Diffusion LLM's Safety Blessing and Its Failure Mode
Zeyuan He, Yupeng Chen, Lang Lin +7
Diffusion large language models (D-LLMs) offer an alternative to autoregressive LLMs (AR-LLMs) and have demonstrated advantages in generation efficiency. Beyond the utility benefit…
Are Targeted Data Poisoning Attacks as Effective as We Think?
William Xu, Chenyu Zhang, Yihan Wang +5
Targeted data poisoning attacks manipulate model predictions on specific test samples by injecting malicious data into training. Yet existing evaluations report average attack succ…
Provable Robust Overfitting Mitigation in Wasserstein Distributionally Robust Optimization
Shuang Liu, Yihan Wang, Yifan Zhu +2
Wasserstein distributionally robust optimization (WDRO) optimizes against worst-case distributional shifts within a specified uncertainty set, leading to enhanced generalization on…
BridgePure: Limited Protection Leakage Can Break Black-Box Data Protection
Yihan Wang, Yiwei Lu, Xiao-Shan Gao +2
Availability attacks, or unlearnable examples, are defensive techniques that allow data owners to modify their datasets in ways that prevent unauthorized machine learning models fr…
MUC: Machine Unlearning for Contrastive Learning with Black-box Evaluation
Yihan Wang, Yiwei Lu, Guojun Zhang +4
Machine unlearning offers effective solutions for revoking the influence of specific training data on pre-trained model parameters. While existing approaches address unlearning for…
Efficient Availability Attacks against Supervised and Contrastive Learning Simultaneously
Yihan Wang, Yifan Zhu, Xiao-Shan Gao
Availability attacks can prevent the unauthorized use of private data and commercial datasets by generating imperceptible noise and making unlearnable examples before release. Idea…